Skol

Skol

About

Skol (for Syzkrash Ordinary Language) is a programming language I've developed for a pretty long time. While it doesn't have one specific purpose, it was a great learning experience to develop. I do not work on the language anymore, but if you wish to continue its legacy, the repository is public. One of the things that'll definitely catch you off-guard is it's syntax, which actually has a pretty long history.

Quick Overview

Skol was written in Go. The source code can be found on GitHub. If you want to learn the language, visit the Skol Documentation.

One of the ways I'd describe Skol is minimalistic. Mostly because I don't want to spend the rest of my life working on it. Another reason is because I wanted to create a language completely free of keywords. The goal was to be able to name your variable, function or structure anything you'd like.

While Skol does mostly achieve this goal, a problem arises with functions. Expressions do not exist in Skol, being replaced by calls to built-in functions. Due to this the general recommendation is to use PascalCase for user-defined functions and snake_case for built-in functions. A problem may also arise when defining structures, as the primitive types' names are not case-sensitive. This would cause confusion between a structure B and the boolean type (which has multiple names: b, bool and boolean)

Backstory

Major influences

If you don't want to know the entire story of how Skol came to be and just want to know what was the biggest influence on its syntax the below paragraph is all you need to read.

The biggest influences are from 3 languages I've designed long before Skol. When it comes to regular programming languages, influences include (but are not limited to) Go, C, and Rust. Code golf was also an influence toward a more concise syntax.

String interpolation

Skol's syntax stemmed off a concept for a string interpolation language I've come up with extremely long ago. Back then, I was merely dreaming of what could be. I can't recall exactly, but the basic syntax looked like this:

Hello %name!

While this seems pretty simple, the language offered more than just referring to predefined name. Namely, function calls:

Hello $capitalize:%name

Quite an odd decision, but that lead me down a rabbit hole: What if there was a syntax to define a function?

$?capitalize[name] ... $ Hello $!capitalize:%!name

At this point this was becoming less and less string interpolation, but just straight up programming.

Datasets

Another language that ended up playing a big influence on Skol's syntax is a dataset language.

Two flavors of the language existed. One for un-keyed data (array data):

somebody once told me

And one for keyed data (dictionary data):

somebody -> once told -> me

Of course, it took inspiration from that string interpolation language, and included references:

@ comments are denoted with the at sign (@) somebody once told me @ produces "somebody once told me" ^$0 ^$1 ^$2 ^$0

The syntax for dictionary data was slightly different:

somebody -> once told -> me sentence -> somebody ^:somebody told ^:told

But I didn't stop there. You could cross-reference values between files.

If we assume "dataset1" contains:

somebody me

Then:

sentence -> ^!dataset1$0 once told ^!dataset1$1

But wait! There's more: Random chance!

@ 50% of the time: @ "somebody once told me" @ the other 50%: @ "somebody once told" sentence -> somebody once told ^?50!dataset1$1

Why would you need random chance in a dataset language? I have no idea!

I never got around to implementing either of these languages. I'm pretty sure the reason is obvious. But the inspiration Skol took from these concepts is pretty clear.

Exk

However, before I implemented Skol, I wanted to implement a language called Exk. It is a direct predecessor of Skol. Here's what a piece of Exk code (would) look like:

exk ?$add_and_print a int; b int { ?%val = !$math::add a; b !$std::println !$std::to_str !%val }

Personally, I'd take Skol's syntax over Exk's any day of the week.

Rough timeline of events